Abstract(#br)In this paper, we provide an optimization framework for channel-adaptive pulse and lattice scaling in a general Time-Frequency (T-F) multiplexing scheme. Closed form relations for the optimum pulse and lattice scaling are derived analytically considering the requirements of two important 5G applications, i.e. tactile internet and cognitive radio. For tactile internet, we solve the problem with a constraint on the time dispersion of the pulse shaping filter at the receiver in order to decrease the length of Cyclic Prefix (CP) and consequently improve the spectral efficiency. Then, aiming for limiting the Out-Of-Band (OOB) leakage in cognitive radio applications, we find a solution to the optimization problem with a constraint on frequency dispersion of pulse shaping filter at... the transmitter. Finally, we consider both of these constraints and solve the problem analytically. The results show that using optimum pulse and lattice scaling, improves the Signal to Interference Ratio (SIR) significantly, compared to non-optimal pulses and lattices. Furthermore, it is shown that applying the constraint to either Receive (Rx-) filter time dispersion or Transmitter (Tx-) filter frequency dispersion does not degrade the SIR significantly while limiting both of them leads to a mismatch between pulse shaping filters and degrades the SIR considerably.